Stereoscopic video quality assessment method based on adaptive weight adjustment

Abstract

The invention belongs to the field of video processing, and provides a three-dimensional video objective quality evaluation method that is in accordance with human visual habits and is more effective. The new three-dimensional video objective quality evaluation method is more accurate and efficient in evaluation of the three-dimensional video quality and promotes development of the three-dimensional imaging technology to a certain degree. The adaptive weight adjustment-based three-dimensional video quality evaluation method comprises a first step of performing weight vector initialization firstly when algorithm begins; a second step of entering an iterative optimization cycle process of the adaptive weight adjustment algorithm, firstly, performing mutation based on the weight vector so as to generate a first generation of weight vector population, wherein the size of the population is N; and a third step of replacing the jth dimensional element of optimal individuals of the previous generation with a positive variance factor and a negative variance factor, and then selecting optimal weight vector individuals of the generation. The method is mainly applied to video processing occasions.

Description

technical field [0001] The invention belongs to the field of video processing, and relates to the improvement and optimization of a stereoscopic video quality evaluation method, in particular to the application of a genetic algorithm and an adaptive weight distribution algorithm in the objective evaluation of stereoscopic video quality. Specifically, it involves a stereoscopic video quality evaluation method based on adaptive weight adjustment. Background technique [0002] Stereoscopic video can enable the viewer to obtain a three-dimensional sense, which undoubtedly enhances the user's viewing experience and makes the user's experience more real. Image processing, display technology and video coding technology (such as H.264 / AVC, H.264 / SVC and multi-view video coding (MVC)) make stereoscopic video technology widely developed in many application fields. Recently, the latest HEVC / H.265 standard [1] Support for stereoscopic video coding has been added. Compared with H.264, ...

AI Technical Summary

Problems solved by technology

Among them, the full reference evaluation method refers to the need for all the original stereoscopic video as a reference when evaluating the objective quality; the partial reference evaluation method refers to the need for some parameters of the original stereoscopic video as a reference when evaluating the objective quality. Based on the fact that the original stereoscopic video is not required, this method can effectively save bandwidth resources compared with the full-reference stereoscopic video quality evaluation method [13] ; The no-reference evaluation method means that the original stereoscopic video is not used as a reference at all when performing objective quality evaluation. This method requires the least transmission cost, but it is also the most difficult to achieve

Images